A Weighted Ensemble Model for Prediction of Infectious Diseases

Abstract

Background: The ensemble building is a common method to improve the performance of the model in case of regression as well as classification. Objective: In this paper we propose a weighted average ensemble model to predict the number of incidence for infectious diseases like typhoid and compare it with applied models for prediction. Methods: The Monthly data of dengue and typhoid cases from 2014 to 2017 were taken from integrated diseases surveillance programme, Government of India. The data was processed by three regressions such as support vector regression, neural network and linear regression. Results: To evaluate the prediction error and compare it with different models, different performance measures have been used such as MSE, RMSE and MAE and it was found that proposed ensemble method performed better in terms of forecast measures. Conclusion: Our main aim in this paper is to minimize the prediction error; the resulting proposed weighted average ensemble model has shown a significant result in terms of prediction errors.

The Prediction of Infectious Diseases: A Bibliometric Analysis

“…An increasing number of the relevant scholars have studied the prediction of infectious diseases and published many papers in the past two decades [ 6 , 7 , 8 ]. To provide a comprehensive review of the prediction of infectious diseases, some scholars have proposed systematic reviews, including Racloz et al [ 9 ], Huppert and Katriel [ 10 ], Christaki [ 11 ], Alessa and Faezipour [ 12 ].…”

Temporal and Spatiotemporal Arboviruses Forecasting by Machine Learning: A Systematic Review

“…When we evaluated the studies regarding the types of models used in the predictions, we observed that the vast majority of authors investigated moving average models (27), such as the Autoregressive Integrated Moving Average (ARIMA) (17,23,29,35,41,43,46,56,(61)(62)(63), Seasonal Autoregressive Integrated Moving Average (SARIMA) (55,(63)(64)(65)(66), Autoregressive Integrated Moving Average with Explanatory Variable (ARIMAX) (67). Several works have also presented a wide variety of models using artificial neural networks, mainly the LSTM (59,(68)(69)(70).…”

“…But models using backpropagation neural networks (BPNN), GANN networks (60), Elman Recurrent Neural Network Levenberg Marquardt Algorithm (ERMN/LMA) (22), and Deep feed-forward neural networks (28) were also investigated. Although neural networks have been extensively explored, in many studies, the authors did not explain the type of network they were investigating (23,45,46,61,66,71,72).…”

OWAE-Net: COVID-19 detection from ECG images using deep learning and optimized weighted average ensemble technique

“…Shashvat et al. Shashvat et al. (2019) predicted the incidence of infectious diseases like typhoid using an ensemble of models like mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE).…”